林业科学 ›› 2023, Vol. 59 ›› Issue (6): 74-87.doi: 10.11707/j.1001-7488.LYKX20220553
焦强英1,2(),韩宗甫2,王炜烨3,刘迪4,潘鹏旭5,李博6,张念慈7,王萍1,陶金花2,范萌2,*
收稿日期:
2022-08-10
出版日期:
2023-06-25
发布日期:
2023-08-08
通讯作者:
范萌
E-mail:jqy19971110@163.com
基金资助:
Qiangying Jiao1,2(),Zongfu Han2,Weiye Wang3,Di Liu4,Pengxu Pan5,Bo Li6,Nianci Zhang7,Ping Wang1,Jinhua Tao2,Meng Fan2,*
Received:
2022-08-10
Online:
2023-06-25
Published:
2023-08-08
Contact:
Meng Fan
E-mail:jqy19971110@163.com
摘要:
目的: 基于长时间序列多源数据,开展雷击火驱动因子分析,采用机器学习方法构建动态、高分辨率雷击火火险预测模型,为雷击火防控提供支撑。方法: 分析大兴安岭地区2010—2020年雷击火时空分布规律,基于闪电监测数据、卫星遥感数据、气象再分析资料、DEM等多源数据选取闪电、气象、植被、地形4类18个雷击火潜在驱动因子,研究其特征及与雷击火发生的关系;提取历史雷击火点和随机生成的非雷击火点对应的驱动因子,构建原始样本集,计算各驱动因子的重要性和相关性矩阵进行驱动因子挑选;基于优化后的训练样本集,采用梯度提升决策树(GBDT)、随机森林(RF)和极端随机树(ERT)3种集成学习模型进行雷击火火险预测能力评估,选择表现最优的方法用于构建大兴安岭地区雷击火火险预测模型并应用。结果: 2010―2020年大兴安岭雷击火出现次数最多和最少的年份分别为2015和2012年,主要集中于5―7月,发生时段主要集中于10:00―17:00;雷击火高发区为漠河县、塔河县、新林区和呼中区。闪电与雷击火的空间分布基本一致,但闪电越多雷击火发生次数不一定越多,2011年闪电次数最多,为114 632次,雷击火仅出现11次。在闪电强度为?20~?40 kA、陡度为?4~?8 kA·μs?1、相对湿度小于40%、降水量小于4 mm、气温大于29 ℃、大气压为91~95 kPa且风速为1~3 m·s–1的气象条件下更易发生雷击火。NDVI(归一化植被指数)、GPP(总初级生产力)、Et(蒸散量)、NPP(净初级生产力)大小与雷击火发生呈正相关。雷击火在海拔300~900 m、坡度0~12°范围内出现次数较多,坡向对雷击火发生影响不大。特征选择后剩余13个特征参量分别参与3种集成学习算法模型构建,其中,ERT模型预测能力最好,其AUC达0.97,查准率、查全率和F1 Score均高于GBDT和RF模型。ERT模型预测的高风险区与实际雷击火点分布区的空间一致性很好。结论: 利用多源大数据,尤其是卫星观测数据,获取更多与雷击火发生有关的潜在驱动因子,并依靠机器学习,能够较好体现因子间非线性关系及自行学习参数间复杂关系能力的优势,本研究构建的雷击火火险预测模型具备很好的泛化性、自适应性和较高的空间分辨率。
中图分类号:
焦强英,韩宗甫,王炜烨,刘迪,潘鹏旭,李博,张念慈,王萍,陶金花,范萌. 基于多源数据和机器学习方法的大兴安岭地区雷击火驱动因子及火险预测模型[J]. 林业科学, 2023, 59(6): 74-87.
Qiangying Jiao,Zongfu Han,Weiye Wang,Di Liu,Pengxu Pan,Bo Li,Nianci Zhang,Ping Wang,Jinhua Tao,Meng Fan. Driving Factors and Forecasting Model of Lightning-Caused Forest Fires in Daxing’ anling Mountains Based on Multi-Sources Data and Machine Learning Method[J]. Scientia Silvae Sinicae, 2023, 59(6): 74-87.
表1
雷击火风险驱动因子属性"
类别 Type | 风险驱动因子 Risk factor | 符号 Symbol | 单位 Unit | 时间分辨率 Temporal resolution | 空间分辨率 Spatial resolution | 数据源 Data source | 是否参与建模 Whether used in modeling |
闪电 Lightning | 引起雷击火的闪电总个数 Lightning-caused fire number | Frq_5 | 次times | 1 d | — | ADTD | 是Y |
Frq_10 | 次times | 1 d | — | ADTD | 否N | ||
引起雷击火的闪电强度均值 Averaged lightning intensity | I_5 | kA | 1 d | — | ADTD | 是Y | |
I_10 | kA | 1 d | — | ADTD | 否N | ||
引起雷击火的闪电陡度均值 Averaged lightning steep | Steep_5 | kA·μs?1 | 1 d | — | ADTD | 否N | |
Steep_10 | kA·μs?1 | 1 d | — | ADTD | 是Y | ||
气象 Meteorology | 相对湿度Relative humidity | Rh | % | 1 h | 0.1° | ERA5 | 是Y |
总降水量 Total precipitation | Tp | m | 1 h | 0.1° | ERA5 | 是Y | |
气温 Temperature | Tm | K | 1 h | 0.1° | ERA5 | 是Y | |
气压 Air pressure | Sp | Pa | 1 h | 0.1° | ERA5 | 是Y | |
风速 Wind speed | Ws | m·s–1 | 1 h | 0.1° | ERA5 | 是Y | |
植被 Vegetation | 归一化植被指数均值 Normalized difference vegetation index | NDVI | — | 16 d | 250 m | MODIS | 是Y |
总初级生产力Gross primary productivity | GPP | g·m?2 | 8 d | 500 m | MODIS | 是Y | |
净初级生产力Net primary productivity | NPP | g·m?2 | 1 a | 1 km | MODIS | 否N | |
蒸散量Evapotranspiration | Et | kg·m?2 | 8 d | 0.04° | MODIS | 是Y | |
地形 Terrain | 高程 Elevation | Ele | m | — | 30 m | SRTM | 否N |
坡度 Slope | Slope | ° | — | 30 m | SRTM | 是Y | |
坡向 Aspect | Aspect | — | — | 30 m | SRTM | 是Y |
表4
2010—2020年大兴安岭地区不同强度等级的闪电次数"
闪电强度 Lightning intensity/kA | 总闪次数 Total lightning frequency | 负闪次数 Negative lightning frequency | 正闪次数 Positive lightning frequency |
0~20 | 477 015(59.20%) | 455 961(60.52%) | 21 054(40.21%) |
20~40 | 224 000(27.80%) | 207 937(27.60%) | 16 063(30.68%) |
40~60 | 64 507(8.01%) | 56 857(7.55%) | 7 650(14.61%) |
60~80 | 21 546(2.67%) | 18 016(2.39%) | 3 530(6.74%) |
>80 | 18 658(2.32%) | 14 590(1.94%) | 4 068(7.77%) |
表5
2010—2020年大兴安岭地区雷击火发生与气象因子的相关性"
指标 Index | 雷击火发生次数 Number of lightning fires | Rh | Tp | t2m | Sp | Ws | NDVI | GPP | NPP | Et |
雷击火发生次数 Number of lightning fires | 1 | |||||||||
Rh | ?0.412 | 1 | ||||||||
Tp | ?0.200 | 0.659 | 1 | |||||||
Tm | 0.456 | ?0.319 | ?0.386 | 1 | ||||||
Sp | 0.003 | ?0.133 | 0.022 | ?0.114 | 1 | |||||
Ws | ?0.095 | ?0.074 | 0.299 | ?0.114 | 0.019 | 1 | ||||
NDVI | 0.372 | 0.347 | 0.183 | 0.518 | ?0.350 | ?0.315 | 1 | |||
GPP | 0.355 | ?0.091 | ?0.083 | 0.443 | ?0.069 | ?0.262 | 0.548 | 1 | ||
NPP | 0.266 | ?0.052 | 0.102 | ?0.036 | ?0.048 | 0.011 | 0.134 | 0.498 | 1 | |
Et | 0.022 | 0.270 | 0.245 | 0.339 | 0.285 | ?0.070 | 0.320 | 0.066 | ?0.107 | 1 |
表7
雷击森林火险等级划分"
火险等级 Fire risk level | 雷击火灾发生概率 The probability of lightning-caused fire occurrence | 预报服务用语 Forecast service conditions |
1级(低火险) Level 1 (low fire risk) | | 雷击火险等级低 The lightning-caused fire risk level is low |
2级(较低火险) Level 2 (moderate low fire risk) | | 雷击火险等级较低 The lightning-caused fire risk level is relatively low |
3级(较高火险) Level 3 (moderate high fire risk) | | 雷击火险等级较高,须加强防范 The risk of lightning-caused fire is high and requires increased prevention measures |
4级(高火险) Level 4 (high fire risk) | | 雷击火险等级高,林区须加强火源管理 The risk of lightning-caused fire is high, and effective fire source management should be strengthened in forested areas |
5级(极高火险) Level 5 (Extremely high fire risk) | | 雷击火险等级极高,严禁 一切林内用火 The risk of lightning-caused fire is extremely high, and all forms of open flames or fires are strictly prohibited within forested areas |
陈宏伟, 胡远满常, 禹 布, 等. 大兴安岭呼中林区虫害与火干扰交互作用的长期模拟. 应用生态学报, 2011, 22 (3): 585- 592. | |
Chen H W, Hu-Yuan M C, Yu B, et al. Interaction between pest and fire disturbances in Huzhong area of Great Xing’ an Mountains: long-term simulation. Chinese Journal of Applied Ecology, 2011, 22 (3): 585- 592. | |
成 勤, 张科杰, 刘 俊, 等. 一次特大暴雨过程三维和二维系统闪电特征对比分析. 热带气象学报, 2021, 37 (3): 396- 408. | |
Cheng Q, Zhang K J, Liu J, et al. Three-dimensional and two-dimensional system lightning in a torrential rain process comparative analysis of characteristics. Journal of Tropical Meteorology, 2021, 37 (3): 396- 408. | |
丁 鹏, 徐爱俊, 周素茵. 基于梯度提升决策树多特征结合的茶叶产量预测. 西南农业学报, 2021, 34 (7): 1556- 1563. | |
Ding P, Xu A J, Zhou S Y. Prediction of tea yield based on gradient boosting decision tree with combining multiple features. Southwest China Journal of Agricultural Sciences, 2021, 34 (7): 1556- 1563. | |
杜春英, 李 帅, 刘 丹, 等. 大兴安岭地区森林雷击火发生的时空分布. 自然灾害学报, 2010, 19 (3): 72- 76. | |
Du C Y, Li S, Liu D, et al. Temporal and spatial distribution of lightning-caused forest fires in Daxing’anling area. Journal of Natural Disasters, 2010, 19 (3): 72- 76. | |
高永刚, 顾 红, 张广英. 大兴安岭森林雷击火综合指标研究. 中国农学通报, 2010, 26 (6): 87- 92. | |
Gao Y G, Gu H, Zhang G Y. Integrated index study on forest lightning fire for Daxing’anling Mountains. Chinese Agricultural Science Bulletin, 2010, 26 (6): 87- 92. | |
郭福涛, 胡海清, 金 森, 等. 基于负二项和零膨胀负二项回归模型的大兴安岭地区雷击火与气象因素的关系. 植物生态学报, 2010, 34 (5): 571- 577. | |
Guo F T, Hu H Q, Jin S, et al. Relationship between forest lighting fire occurrence and weather factors in Daxing’an Mountains based on negative binomial model and zero-inflated negative binomial models. Chinese Journal of Plant Ecology, 2010, 34 (5): 571- 577. | |
何 诚, 舒立福, 刘柯珍. 大兴安岭地区夏季森林火灾环境因子特征分析. 西南林业大学学报(自然科学), 2021, 41 (3): 87- 93. | |
He C, Shu L F, Liu K Z. Analysis on environmental factors’ characteristics of summer forest fire in Daxing’an Mountains. Journal of Southwest Forestry University(Natural Science), 2021, 41 (3): 87- 93. | |
贺 敬, 张庚辉, 王国军. 小兴安岭森林火灾发生规律及林火天气等级模型的建立. 黑龙江气象, 2011, 28 (2): 21- 22. | |
He J, Zhang G H, Wang G J. Occurrence regularity of forest fire in Xiaoxing ' an Mountains and establishment of forest fire weather grade model. Heilongjiang Meteorology, 2011, 28 (2): 21- 22. | |
胡海清, 魏书精, 孙 龙. 大兴安岭2001-2010年森林火灾碳排放的计量估算. 生态学报, 2012, 32 (17): 5373- 5386.
doi: 10.5846/stxb201109091324 |
|
Hu H Q, Wei S J, Sun L. Estimating carbon emissions from forest fires during 2001 to 2010 in Daxing’anling Mountain. Acta Ecologica Sinica, 2012, 32 (17): 5373- 5386.
doi: 10.5846/stxb201109091324 |
|
刘柯珍, 赵凤君, 王明玉, 等. 基于零膨胀负二项模型对大兴安岭地区夏季火发生预报模型的研究. 林业机械与木工设备, 2018, 46 (6): 7- 11. | |
Liu K Z, Zhao F J, Wang M Y, et al. ZINB-based research on a model for forecasting occurrence of forest fires in Daxing’anling area in summer. Forestry Machinery & Woodworking Equipment, 2018, 46 (6): 7- 11. | |
李 强, 史海峰. 基于ADTD与VLF/LF闪电监测系统的数据对比分析. 科技通报, 2021, 37 (11): 32- 36, 84. | |
Li Q, Shi H F. Comparative analysis of datas based on ADTD and VLF/LF lightning location system. Bulletin of Science and Technology, 2021, 37 (11): 32- 36, 84. | |
马文苑, 冯仲科, 成竺欣, 等. 山西省林火驱动因子及分布格局研究. 中南林业科技大学学报, 2020, 40 (9): 57- 69. | |
Ma W Y, Feng Z K, Cheng Z X, et al. Study on driving factors and distribution pattern of forest fires in Shanxi province. Journal of Central South University of Forestry & Technology, 2020, 40 (9): 57- 69. | |
倪长虹, 邸雪颖. 黑龙江省大兴安岭雷击火发生规律. 东北林业大学学报, 2009, 37 (1): 55- 57,75. | |
Ni C H, Di X Y. Occurrence regularity of lightning fire in Daxing’anling, Heilongjiang. Journal of Northeast Forestry University, 2009, 37 (1): 55- 57,75. | |
彭 欢, 史明昌, 孙 瑜, 等. 基于Logistic的大兴安岭雷击火预测模型. 东北林业大学学报, 2014, 42 (7): 166- 169. | |
Peng H, Shi C M, Sun Y, et al. Lightning fire forecasting model of Daxing’an Mountain based on logistic model. Journal of Northeast Forestry University, 2014, 42 (7): 166- 169. | |
舒立福, 田晓瑞, 李 骞, 等. 雷击火形成及预防扑救对策. 火灾科学, 2014, 8 (3): 31- 37. | |
Shu L F, Tian X R, Li Q, et al. 1999. The formation of lightning fire and its prevention countermeasures. Fire Safety Science, 2014, 8 (3): 31- 37. | |
舒 洋, 孙子瑜, 张 恒. 世界森林雷击火研究现状和展望. 世界林业研究, 2022, 35 (2): 34- 40. | |
Shu Y, Sun Z Y, Zhang H. Current status and prospects of research on lightning strike in forest fire. World Forestry Research, 2022, 35 (2): 34- 40. | |
苏漳文. 2020. 基于地理信息系统的大兴安岭林火发生驱动因子及预测模型的研究. 哈尔滨: 东北林业大学. | |
(Su Z W. 2020. Study on driving factors and forecasting models of forest fire in the Xing’an Mountains based on geographic information system. Harbin: Northeast Forestry University.[in Chinese]). | |
孙 瑜, 史明昌, 彭 欢, 等. 基于MAXENT模型的黑龙江大兴安岭森林雷击火火险预测. 应用生态学报, 2014, 25 (04): 1100- 1106. | |
Sun Y, Shi M C, Peng H, et al. Forest lighting fire forecasting for Daxing’anling Mountains based on MAXENT model. Chinese Journal of Applied Ecology, 2014, 25 (04): 1100- 1106. | |
田晓瑞, 舒立福, 赵凤君, 等. 大兴安岭雷击火发生条件分析. 林业科学, 2012, 48 (7): 98- 103. | |
Tian X R, Shu L F, Zhao F J, et al. Analysis of the conditions for lightning fire occurrence in Daxing' anling region. Scientia Silvae Sinicae, 2012, 48 (7): 98- 103. | |
王会福, 龚立群. 秦岭林区雷击火的发生和预防. 陕西林业科技, 2013, 72 (5): 69- 70. | |
Wang H F, Gong L Q. Occurrence and prevention of lightning fire in Qinling forest region. Shanxi Forest Science and Technology, 2013, 72 (5): 69- 70. | |
王晓红, 黄 艳, 张吉利, 等. 基于闪电定位数据和气象数据的大兴安岭雷击火预测模型研究. 中南林业科技大学学报, 2017, 37 (3): 44- 48. | |
Wang X H, Huang Y, Zhang J L, et al. Research on daily prediction model of lightning fires in Daxing’an ling region based on lightning location data. Journal of Central South University of Forestry & Technology, 2017, 37 (3): 44- 48. | |
韦良芳. 2020. 基于极端随机树与Logistic回归算法的网贷平台个人信用评估模型的比较研究. 济南: 山东大学. | |
(Wei L F. 2020. A comparative study of personal credit evaluation model on net lending platforms based on extremely randomized trees with logistic regression algorithm. Jinan: Shandong University.[in Chinese]). | |
于建龙, 刘乃安. 我国大兴安岭地区森林雷击火发生的火险天气等级研究. 火灾科学, 2010, 19 (3): 131- 137. | |
Yu J L, Liu N A. Study on the fire dangerous weather level of forest thunder fire in Daxing’an Mountains of China. Fire Safety Science, 2010, 19 (3): 131- 137. | |
臧桐汝, 舒立福, 王明玉, 等. 黑龙江大兴安岭林区雷击火时空分布及驱动因素分析. 西北农林科技大学学报(自然科学版), 2022, 50 (12): 2- 14. | |
Zang T R, Shu L F, Wang M Y, et al. Analysis of spatial and temporal distribution and driving factors of lightning-caused fires in Daxing’anling forest region of Heilongjiang. Journal of Northwest A & F University(Natural Science Edition), 2022, 50 (12): 2- 14. | |
张吉利, 毕 武, 王晓红, 等. 雷击火发生的影响因子与预测研究进展. 应用生态学报, 2013, 24 (9): 2674- 2684. | |
Zhang J L, Bi W, Wang X H, et al. Lightning-caused fire, its affecting factors and prediction: A review. Chinese Journal of Applied Ecology, 2013, 24 (9): 2674- 2684. | |
张 媛, 李胜男, 张运生. 森林雷击火特点和监测预警技术研究进展. 森林防火, 2018, 36 (3): 44- 48. | |
Zhang Y, Li S N, Zhang Y S. Research progress of forest thunder fire characteristics and monitoring and early warning technology. Forest Fire Prevention, 2018, 36 (3): 44- 48. | |
赵鲁强, 杨晓丹, 梁 莉, 等. 2018. GB/T 36743—2018森林火险气象等级. 北京: 中国标准出版社. | |
Zhao L Q, Yang X D, Liang L, et al. 2018. GB/T 36743—2018 Forest fire danger weather ratings. Beijing: Standards Press of China.[in Chinese] | |
Albert-Green A, Dean C B, Martell D L, et al. A methodology for investigating trends in changes in the timing of the fire season with applications to lightning-caused forest fires in Alberta and Ontario, Canada. Canadian Journal of Forest Research-revue Canadienne De Recherche Forestiere, 2013, 43 (1): 39- 45.
doi: 10.1139/cjfr-2011-0432 |
|
Barbero R, Abatzoglou J T, Kolden C A, et al. Multi-scalar influence of weather and climate on very large-fires in the Eastern United States. International Journal of Climatology, 2015, 35 (8): 2180- 2186.
doi: 10.1002/joc.4090 |
|
Chen F, Du Y, Niu S, et al. Modeling forest lightning fire occurrence in the Daxinganling Mountains of northeastern China with MAXENT. Forests, 2015, 6 (5): 1422- 1438. | |
Dickson B G. , Prather J W, Xu Y, et al. 2006. Mapping the probability of large fire occurrence in northern Arizona, USA. Landscape Ecology. 21(5): 747-761. | |
Friedman J H. Greedy function approximation: a gradient Boosting machine. Annals of Statistics, 2001, 29 (5): 1189- 1232.
doi: 10.1214/aos/1013203450 |
|
Geurts P, Ernst D, Wehenkel L. Extremely randomized trees. Machine Learning, 2006, 63 (1): 3- 42.
doi: 10.1007/s10994-006-6226-1 |
|
Guo F, Wang G, Su Z, et al. What drives forest fire in Fujian, China? Evidence from logistic regression and random forests. International Journal of Wildland Fire, 2016, 25 (5): 505.
doi: 10.1071/WF15121 |
|
Krenn R, Hergarten S. Cellular automaton modelling of lightning-induced and man made forest fires. Natural Hazards Earth System Sciences, 2009, 9 (5): 1743- 1748.
doi: 10.5194/nhess-9-1743-2009 |
|
Miyoshi S, Uezu T, Okada M. Statistical mechanics of time domain ensemble learning. Journal of the Physical Society of Japan, 2006, 75 (8): 2652- 2674. | |
Pew K L, Larsen C P S. GIS analysis of spatial and temporal patterns of human-caused wildfires in the temperate rain forest of Vancouver Island, Canada. Forest ecology and management, 2001, 140 (1): 1- 18.
doi: 10.1016/S0378-1127(00)00271-1 |
|
Scornet E. Random forests and kernel methods. IEEE Transactions on Information Theory, 2015, 62 (3): 1485- 1500. | |
Shevyrnogov A P, Botvich I Y, Kononova N A, et al. Remote ground-based and satellite monitoring of vegetation. Herald of the Russian Academy of Sciences, 2019, 88 (6): 469- 474. | |
Wang S, Li H L, Niu S K. Empirical research on climate warming risks for forest fires: A case study of grade I forest fire danger zone, Sichuan province, China. Sustainability, 2021, 13 (14): 7773.
doi: 10.3390/su13147773 |
|
Wu Z W, He H S, Yang J, et al. Relative effects of climatic and local factors on fire occurrence in boreal forest landscapes of northeastern China. Science of the Total Environment, 2014, 493 (5): 472- 480. | |
Zhang Y G, Ju W M, Migliavacca M, et al. Advances in hyperspectral remote sensing of vegetation traits and functions. Remote Sensing of Environment, 2021, 252, 112121.
doi: 10.1016/j.rse.2020.112121 |
[1] | 周小成, 黄婷婷, 李媛, 肖祥希, 朱洪如, 陈芸芝, 冯芝淸. 结合遥感林龄因子的亚热带森林蓄积量估算方法[J]. 林业科学, 2023, 59(4): 88-99. |
[2] | 谭炳香,沈明潭,郄光发,戚瞾,贺晨瑞. 冬奥会崇礼生态核心区植被覆盖时空变化遥感监测[J]. 林业科学, 2022, 58(4): 141-151. |
[3] | 杨楠楠,白岩,姜苏译,杨春梅,徐凯宏. 基于AlexNet优化的板材材种识别方法[J]. 林业科学, 2022, 58(3): 149-158. |
[4] | 司莉青,苑尚博,赵凤君,舒立福,王明玉,马启明. 雷电、雷击火发生与人类活动关系[J]. 林业科学, 2022, 58(11): 1-9. |
[5] | 王明玉,苑尚博,李威,李伟克,宋佳军,司莉青,王亚惠,赵凤君,田晓瑞,李笑笑,舒立福. 密集雷电引发大兴安岭群发雷击火过程及其影响因素[J]. 林业科学, 2022, 58(11): 10-20. |
[6] | 李伟克,舒立福,苑尚博,宋佳军,李威,司莉青,赵凤君,王亚惠,王明玉. 基于VLF/LF三维闪电定位系统的大兴安岭闪电时空分布特征[J]. 林业科学, 2022, 58(11): 21-30. |
[7] | 李贾,兰岚,张佐忠,袁文涛,高德民,宗树琴,业巧林. 基于深度学习的森林可燃物含水率反演技术[J]. 林业科学, 2022, 58(10): 47-58. |
[8] | 周小成,郑磊,黄洪宇. 基于多特征优选的无人机可见光遥感林分类型分类[J]. 林业科学, 2021, 57(6): 24-36. |
[9] | 潘颖,丁鸣鸣,林杰,代侨,郭赓,崔琳琳. 基于PROSAIL模型和多角度遥感数据的森林叶面积指数反演[J]. 林业科学, 2021, 57(4): 90-106. |
[10] | 孙龙,刘祺,胡同欣. 森林地表死可燃物含水率预测模型研究进展[J]. 林业科学, 2021, 57(4): 142-152. |
[11] | 张会儒,雷相东,李凤日. 中国森林经理学研究进展与展望[J]. 林业科学, 2020, 56(9): 130-142. |
[12] | 谢锦莹,丁丽霞,王志辉,刘丽娟. 基于FCN与面向对象的滨海湿地植被分类[J]. 林业科学, 2020, 56(8): 98-106. |
[13] | 赵霖,张晓丽,吴艳双,张斌. 面向机载高光谱数据的3D-CNN亚热带森林树种分类[J]. 林业科学, 2020, 56(11): 97-107. |
[14] | 杜一尘, 李明泽, 范文义, 王斌. 基于地理加权回归模型与林火遥感数据估算森林年龄[J]. 林业科学, 2019, 55(6): 184-194. |
[15] | 李霓雯, 张晓丽, 张凝, 朱程浩, 孙振峰. 基于加权信息量模型的油松毛虫灾害发生危险性评价[J]. 林业科学, 2019, 55(3): 106-117. |
阅读次数 | ||||||
全文 |
|
|||||
摘要 |
|
|||||